-
930
On the design principles of peptide–drug conjugates fortargeted
drug delivery to the malignant tumor siteEirinaios I. Vrettos1,
Gábor Mező2,3 and Andreas G. Tzakos*1
Review Open AccessAddress:1University of Ioannina, Department of
Chemistry, Section of OrganicChemistry and Biochemistry, Ioannina,
GR-45110, Greece, 2EötvösLoránd University, Faculty of Science,
Institute of Chemistry,Pázmány P. stny. 1/A, H-1117 Budapest,
Hungary and 3MTA-ELTEResearch Group of Peptide Chemistry, Hungarian
Academy ofSciences, Eötvös Loránd University, Pázmány P. stny. 1/A,
H-1117Budapest, Hungary
Email:Andreas G. Tzakos* - [email protected]
* Corresponding author
Keywords:bioconjugates; cancer; drug delivery; PDC; peptide;
peptide–drugconjugate; side-products in PDCs
Beilstein J. Org. Chem. 2018, 14,
930–954.doi:10.3762/bjoc.14.80
Received: 29 January 2018Accepted: 04 April 2018Published: 26
April 2018
This article is part of the Thematic Series "Peptide–drug
conjugates".
Guest Editor: N. Sewald
© 2018 Vrettos et al.; licensee Beilstein-Institut.License and
terms: see end of document.
AbstractCancer is the second leading cause of death affecting
nearly one in two people, and the appearance of new cases is
projected to riseby >70% by 2030. To effectively combat the
menace of cancer, a variety of strategies have been exploited.
Among them, the devel-opment of peptide–drug conjugates (PDCs) is
considered as an inextricable part of this armamentarium and is
continuouslyexplored as a viable approach to target malignant
tumors. The general architecture of PDCs consists of three building
blocks: thetumor-homing peptide, the cytotoxic agent and the
biodegradable connecting linker. The aim of the current review is
to provide aspherical perspective on the basic principles governing
PDCs, as also the methodology to construct them. We aim to offer
basic andintegral knowledge on the rational design towards the
construction of PDCs through analyzing each building block, as also
to high-light the overall progress of this rapidly growing field.
Therefore, we focus on several intriguing examples from the recent
litera-ture, including important PDCs that have progressed to phase
III clinical trials. Last, we address possible difficulties that
mayemerge during the synthesis of PDCs, as also report ways to
overcome them.
930
IntroductionCurrent cancer chemotherapyCancer is one of the
leading causes of death globally behind theheart and circulatory
disorders based on statistics of WorldHealth Organization (WHO)
[1]. Among all different types of
cancer, the most fatal for males are lung and prostate
cancer,while for females are breast cancer, colon & rectum
cancer [1].Notably, more than 12 million cancer cases and 7
million
https://www.beilstein-journals.org/bjoc/about/openAccess.htmmailto:[email protected]://doi.org/10.3762%2Fbjoc.14.80
-
Beilstein J. Org. Chem. 2018, 14, 930–954.
931
Figure 1: Conventional chemotherapy versus targeted
chemotherapy. Black color = Solid malignant tumor; red =
conventional untargeted cytotoxicagent; blue = targeted cytotoxic
agent.
cancer deaths are estimated to have occurred both in males
andfemales in 2008 worldwide [2]. These numbers have mountedup to
15 million cases and 8.8 million deaths in 2015. Thesestatistics
clearly indicate that cancer is not retreating but iscreeping up,
tending to become the leading cause of mortality.Thus, it can be
concluded that the current therapeutic formula-tions utilized in
oncology are not adequately effective againstthe complexity of
cancer, mostly due to the associated collat-eral toxicity present
in healthy tissues. It is estimated that about30% of the clinical
trials on ClinicalTrials.gov are related tocancer, while only 10%
of them eventually gain marketapproval [3], rendering the drug
development, especially in thistherapeutic direction, costly and
inefficient. Specifically, 12cancer drugs were approved by the FDA
in 2017 [4], com-prising 26% of the total amount of approvals with
respect toother therapeutic areas. These figures suggest that it is
of greatimportance to turn the focus of the global market on
targetedtherapies. In 2009, the total earnings in the United
States,derived from targeted cancer drugs, have reached $10.4
billion,showing an almost 2.2-fold increase since 2005 [5].
However,despite the significant attention that field has gained the
pastdecades, it still remains unfulfilled.
Current treatment processes involve a combination of
surgicalintervention, radiation and chemotherapy. Drugs used for
thispurpose are inevitably cytotoxic in order to eliminate
cancercells, but they lack selectivity that could be developed
throughtargeting malignant cells (Figure 1). Due to the
uncontrolledperipheral toxicity, anticancer drugs usually kill
healthy tissues,resulting in severe effects on the patient’s
health. One represen-tative example is gemcitabine, which
demonstrates higher toxic-ity for healthy cells, after long-term
administration, with respect
to cancer cells. This happens since cancer cells evolve
morerapidly and develop drug resistance by diminishing
expressednucleoside receptors responsible for the cell uptake of
gemcita-bine [6].
Additionally, chemotherapy with anticancer agents is
oftenhampered by their poor aqueous solubility, low oral
bioavail-ability and metabolic instability. These drawbacks are
linked tothe unfavorable ADME (absorption distribution
metabolismexcretion) that are basically described in the following
fourconsecutive axes: 1) Absorption is directly connected with
thetransportation process of the drug from the site of
administra-tion to the systemic circulation [7]. 2) Distribution
refers to thedelivery of the drug to the tissues which usually
occurs via thebloodstream. Conventional chemotherapeutic drugs
(gemcitabi-ne, paclitaxel, doxorubicin, etc.) are not capable to be
selec-tively delivered to the tumor sites and they end up scattered
inthe whole body. 3) Metabolism is a standard biological
strategyfor detoxification, breaking down of the administrated
drugs,once inserted into the human body. The drugs get
decomposedand converted to their metabolites. These metabolites can
bepharmacologically inactive, e.g., gemcitabine converted to
2',2'-difluorodeoxyuridine (dFdU) [8] or possess enhanced
activitywith respect to the parent drug, e.g., temozolomide
converted to5-(3-methyltriazen-1-yl)imidazole-4-carboxamide (MTIC)
[9].4) Excretion is the final step and is responsible for the
removalof the parent drug and/or its metabolites from the human
body.Renal excretion is the predominant route of elimination,
occur-ring via urine.
Therefore, most conventional cytotoxic agents applied
inchemotherapy lack optimum pharmacokinetic properties
-
Beilstein J. Org. Chem. 2018, 14, 930–954.
932
(ADME) and thus are not very effective to treat
malignancies.Moreover, despite the intensive research to construct
new cyto-toxic drugs, survival rates in most cancers remain low
[10] andclinical attrition rates in oncology have been devastating
[11].These data render obvious that the currently developeddrugs,
as also the continuous attempt to discover newones, have not
provided the expected therapeutic impact inoncology.
It is clear that we do have access to an enormous pool of
unspe-cific cytotoxic agents that can efficiently kill cancer
cells. Whatis currently needed is not to invest so intensively in
generatingmore cytotoxic agents but to re-use and re-cycle
available onesand tailor them to be transformed into targeted
chemotherapeu-tics. Along these lines, drug delivery vehicles that
can betailored for different types of cancer and shape
personalizedtherapeutics are continuously gathering attention. Such
drugdelivery systems are of ultimate importance to
effectivelysurpass these hurdles and eventually improve drug
potency.
Charting the malignant tumormicroenvironmentIn order to
selectively deliver cytotoxic drugs to malignanttumor sites,
scientists can take advantage and map first thedifferential
microenvironment between cancer and normal cells.The first one to
report a fundamental difference between malig-nant and normal cells
was Otto Heinrich Warburg in the early1900s, who was awarded the
Nobel Prize in 1931. He proposedthat malignant tumor growth relies
on aerobic glycolysis, incontrast to normal cells that generate
energy by mitochondrialoxidative phosphorylation. The fact that
cells converted pyru-vate to lactate, even in the presence of
oxygen, rendered his ob-servation puzzling for scientists, who
still struggle to elucidatethe complete mechanism of action of
diseased cells. Followingthe Warburg effect, 18F-deoxyglucose
positron emission tomog-raphy (FDG–PET) imaging was developed in
order to visualizethe phenomenon of increased glucose uptake by
cancer cells[12].
Nowadays, it has been demonstrated that malignant cells
differmarkedly in many metabolic aspects compared to normal
cells[13], thus offering the opportunity to target them in
variousways. Most cancer tissues exhibit the following
characteristicsthat can be exploited for developing targeted
cytotoxic agents:
1. Dysregulation of translation initiation factors and
regula-tors [14].
2. Mutations in epigenetic regulatory genes [15].3.
Overexpression of surface receptors like HER2R [16],
folate receptor [17], GnRH receptor [18,19] and aminoacid
transporters [20].
4. Overwhelming production of stimulus agents and en-zymes [21].
For instance, many types of cancer show en-hanced levels of
reactive oxygen species (ROS) whichare reactive molecules and play
a crucial role in cellproliferation [22].
5. The slightly acidic pH of the tumor microenvironment[23]
(Warburg effect).
These are some noteworthy differences that underlie the
dis-crimination between cancerous and normal cells and are
oftenexploited in order to control the site of the drug release
duringtargeted cancer chemotherapy.
ReviewStrategies for targeted delivery of toxicwarheads to
malignant tumor sitesThe main challenge of the drug delivery
concept is to transportsufficient amount of the cytotoxic agent to
a specific locationwith minimum adverse side effects. To conquer
this, various ap-proaches are being exploited at the moment. These
include, butare not limited to: a) utilization of drug delivery
vehicles andformulates like nanoparticles [24] and calixarenes or
cyclo-dextrins [25,26], where the cytotoxic drug is loaded and can
bereleased at the malignant tumor site; b) installation of
labilechemical groups to the tumor microenvironment (i.e., low
pH)able to mask the cytotoxic drug and form a prodrug with
en-hanced plasma stability and/or cell permeability [27] and in
thesame time diminished toxicity for normal cells; c)
covalentattachment of a drug on a tumor-targeting element (small
mole-cule, peptide or antibody) able to selectively target
andpermeate cancer cells. The conjugation is being conducted via
arationally designed linker able to release the drug inside
thecancer microenvironment [19].
The ideal targeting molecular device would consist of
thefollowing modules: a) the cytotoxic agent (drug), b) the
trans-porting - drug delivery vehicle (i.e., lipid, mannan
[28-30]),c) the linker tethering the transporting vehicle to the
cytotoxicwarhead, d) the “programmable” navigating/targeting
moiety(i.e., receptor-specific ligand) and e) the “stealth” carrier
(i.e.,PEG) transfusing enhanced bioavailability. These modules
areencoded in Figure 2A with different colors: the
transportingvehicle in green color, the drug in blue color, the
linker in redcolor, the navigating/targeting agent in black color
and the“stealth” carrier in grey color [31]. The specific color
codingwill be followed, for simplicity purposes, in all examples
oftargeting devices that will be presented throughout this
review.
Among the most intriguing navigating delivery systems that
cancombine the transporting vehicle and the
navigating/targetingmoiety in a single module are the tumor-homing
peptides [32].
-
Beilstein J. Org. Chem. 2018, 14, 930–954.
933
Figure 2: A. General structural architecture of the ideal
navigated drug delivery system [31]. B. General structure of a
peptide–drug conjugate (PDC).
These peptides are exploited to assemble the
peptide–drugconjugates (PDCs) which are considered as prodrugs, due
to thecovalent coupling of a peptide to a drug via specific
linkers. Themain building blocks of a simple PDC include a
cytotoxic agent(drug), a tumor-homing peptide (navigating/targeting
moiety)and a linker between them (Figure 2B).
This class of prodrugs is continuously gaining attention
sincepeptides can be easily produced in large quantities and
theirpurification is simple. Moreover, an array of different
tumor-targeting peptides has been discovered [32] for
multifarioustypes of cancer. This bountiful palette can permit the
construc-tion of personalized cancer therapeutics upon selecting a
tumor-homing peptide that will be most appropriate for the type
ofcancer needed. In addition, peptide sequences can be
selectedaccording to the required physicochemical properties such
assolubility, stability and overall charge or the
characteristicgroups necessary for the conjugation with the
therapeuticpayload. The overall experimental procedure to
synthesize aPDC is usually rapid and facile. Notably, the overall
cost toproduce a PDC, where an already approved drug can beselected
and re-used from a pool of available cytotoxic agents,is much lower
compared to the cost of synthesizing a new cyto-toxic agent, as it
is based on an already applied drug with theaddition of a small
peptide. Nevertheless, the last years morecomplex bioconjugates
have been synthesized to allow the si-multaneous diagnosis and
therapy (theranostics) of diseases.
The therapeutic efficacy of a PDC is predominantly
associatedwith the potency of the drug and the targeting efficiency
of theassembled conjugate. Thus, PDCs should possess certain
fea-tures to render them appealing candidates for treatment:
1. The peptide contained in the PDC must bind selectivelyand
with the optimal affinity to a certain receptor,
present on the cell surface of the targeted tissues and
notwithin their cytosol or nucleus (i.e., steroid
receptors[33]).
2. The selected receptor should be uniquely expressed
oroverexpressed on cancer cells (usually 3-fold or higher
incomparison with normal cells). Additionally, it should
beexpressed in sufficient levels to pump inside the cell
effi-cacious doses of the drug.
3. The peptide-carrier should be constructed in such waythat the
conjugation with a drug or/and a fluorophore isfeasible.
Conjugation usually occurs on lysine, cysteineand glutamic acid
[34] via orthogonal coupling or on thefree N-terminus of the
peptide during solid phase peptidesynthesis. Though, the
conjugation site should be care-fully selected, since perturbations
induced in the peptidestructural microenvironment may result in the
abolish-ment of its binding affinity/selectivity to the targeted
re-ceptor.
4. The linker should be carefully selected to succeed theoptimal
performance of the PDC. An injudicious selec-tion may cause
diminished binding affinity of the peptideto the receptor or/and
reduction of the therapeuticwindow of the drug. Additionally, it
should be enzymati-cally stable during the blood circulation in
order to effi-ciently reach the malignant tumor site and release
thepayload in its microenvironment, reducing the
off-targettoxicity.
5. The cytotoxic agent should contain proper functionalgroup
that can be linked to the tumor homing peptide(i.e., gemcitabine
[19]) or if it is not present it should berationally installed
taking into consideration the final de-rivative of the cytotoxic
agent to retain the original cyto-toxic activity. The sections
below summarize the basicdesign principles of peptide–drug
conjugates to selec-tively target the malignant cells.
-
Beilstein J. Org. Chem. 2018, 14, 930–954.
934
Selecting the proper tumor-targeting peptideto generate the
PDCsThere is an immense variety of peptides (linear or cyclic)
thathave been exploited as carriers/targeting elements to
successful-ly deliver the cytotoxic warhead to cancer cells [32].
Thesepeptides are cell-specific and bind to certain
receptorspromoting their internalization. They are usually inserted
intothe cell via endocytosis and then they are transported to
intra-cellular compartments with higher concentration of enzymesand
lower values of pH, where they disassociate from the recep-tor and
afterward from the anticancer agent. The most represen-tative
examples of peptides utilized for PDCs are highlightedbelow.
Linear peptides are included among the rich reservoir ofoptions,
finding applications in tumor targeting. They exist indifferent
lengths, structures and with various physicochemicalproperties.
Attempting to ameliorate certain disadvantages of linearpeptides
like fast renal clearance or low binding selectivity/affinity due
to the unstable structure of the linear peptides,cyclic peptides
have been introduced. An immense number ofcyclic peptides have been
synthesized [35-37] and many ofthem have displayed superior
affinity and selectivity for the re-ceptor than their parent linear
counterparts [38]. Cyclic peptidesare usually synthesized by
reacting the N-terminus with theC-terminus or by exploiting
specific functional groups ofcertain amino acids present in the
sequence. A representativeexample is the sulfhydryl group of
cysteine-containing peptideswhich may cyclize through the formation
of intramoleculardisulfide bonds [39].
The most commonly used linear peptides and cyclic peptidesthat
can be delivered inside cancer cells via endocytosis and onethat
smuggles into glioma tissues via transcytosis (angiopep-2)are
presented below:
Arginine-glycine-aspartic acid (RGD): A widely appliedpeptide
carrier is the tripeptide arginine-glycine-aspartic acid(RGD)
motif, which was first identified by Ruoslahti andPierschbacher in
the early 1980s [40] within fibronectin thatmediates cell
attachment and was known to target integrin α5β1[41]. In general,
the ‘integrin’ nomenclature was first used in1987 to describe a
family of receptors, appearing asheterodimers of noncovalently
associated α and β subunits, ableto link the extracellular matrix
(ECM) with the intracellularcytoskeleton to mediate cell adhesion,
migration and prolifera-tion [42]. The RGD motif is contained in
various proteins likefibrinogen, fibronectin, prothrombin, tenascin
and other glyco-proteins [43] and is known to be recognized by over
10 inte-
grins, among the over 24 known integrins [44,45], including
allαv integrins, α5β1, α8β1 and αIIbβ3 [46].
The fact that carcinogenesis is highly dependent on
migration,invasion and angiogenesis renders integrins important
anti-cancer targets. Integrin αvβ3 is an important factor in
tumorangiogenesis and metastasis [45], two common characteristicsof
cancer that discriminates it from other diseases. Notably,integrin
αvβ3 (also known as the vitronectin receptor) appearsto be the most
important among all integrins regarding cellproliferation, invasion
and angiogenesis [47]. This integrin isoverexpressed on activated
endothelial cells, new-born vesselsand other tumor cells [48,49],
but it is found to be expressed atundetectable levels in most adult
epithelial cells, making it asuitable target for anti-angiogenic
therapy [50]. Due to its highlevels of expression in cancer cells,
several peptides containingthe RGD motif have been exploited for
the formulation ofPDCs, with the most representative example to be
the peptideCDCRGDCFC [46,51,52].
Gonadotropin-releasing hormone (GnRH): Gonadotropin-releasing
hormone (GnRH), also known as luteinizing hormone-releasing hormone
(LHRH), is a hormone responsible for thesecretion of two
gonadotropins: follicle-stimulating hormone(FSH) and luteinizing
hormone (LH) from the anterior pituitarygland. GnRH is synthesized
and released from GnRH neuronswithin the hypothalamus and
selectively binds to its receptor(GnRH-R), a seven-transmembrane
G-protein-coupled receptor.The structure of the GnRH hormone
(pGlu-His-Trp-Ser-Tyr-Gly-Leu-Arg-Pro-Gly-NH2) was first discovered
in 1971 byBaba et al. [53]. Besides this form, there is GnRH-II
(pGlu-His-Trp-Ser-His-Gly-Trp-Tyr-Pro-Gly-NH2) discovered in
mostvertebrates as well as in humans [54]. This peptide acts
througha similar receptor (type II GnRH-R), which is expressed in
dif-ferent tissues, including tumor cells. Another natural isoform
ofGnRH is GnRH-III (pGlu-His-Trp-Ser-His-Asp-Trp-Lys-Pro-Gly-NH2),
which has been isolated from sea lamprey. GnRH-IIIbinds to GnRH-R
overexpressed on the cancer cell surface, re-sulting in an
antiproliferative effect but seems to be less potentthan the rest
GnRH analogs regarding stimulating gonadotropinrelease at the
pituitary level [55].
GnRH peptide analogs constitute an emerging class of tumorhoming
peptides for malignant tissues expressing the GnRH-R.Their
development is based on the fact that specific humancancer cells
(mostly ovarian, prostate, lung and breast) uniquelyexpress or
overexpress GnRH-R with respect to normal tissues[55-57].
Therefore, covalent attachment of a cytotoxic agent tothese
peptides provides the possibility to produce potent tumor-targeting
PDCs. Various amino acid alterations have been per-formed with
respect to the native hormone [58], while the most
-
Beilstein J. Org. Chem. 2018, 14, 930–954.
935
frequently used GnRH analog is D-Lys6-GnRH-I, which isknown to
bind selectively to GnRH-R. The substitution of Gly6
of the native hormone with D-Lys6 provided an analog withhigher
binding affinity, stabilized β-bend and resistance toproteolytic
cleavage. Moreover, the side chain of lysinecontains a free amine
group (εNH2) allowing orthogonal cou-pling with a cytotoxic warhead
[19]. A considerable number ofPDCs based on GnRH [59-63] exist and
our group has exploitedthis peptide to construct two PDCs
[18,19].
Somatostatin (SST): Somatostatin is a neuropeptide producedby
neuroendocrine, inflammatory and immune cells and has animportant
role in various physiological functions acting as aclassical
endocrine hormone, a paracrine regulator or a neuro-transmitter
[64]. Somatostatin appears in two distinct activeforms:
somatostatin-14 (SST-14) and somatostatin-28 (SST-28).Both SST-14
and SST-28 exhibit biological activity throughhigh-affinity
membrane receptors (somatostatin receptor 1–5;SSTR1–5), that are
widely distributed throughout the humanbody in various tissues like
the nervous, pituitary, kidney, lungand immune cells [65,66].
SSTRs are overexpressed in various neuroendocrine
malignanttumors (NETs) including pancreatic, pituitary, prostate,
lungcarcinoids, osteosarcoma etc. and other non-NETs
includingbreast, colorectal, ovarian, cervical etc. [67].
Therefore, thesereceptors can be targeted for selective delivery of
efficient con-centrations of cytotoxic warheads to the tumor sites.
However,native somatostatin gets rapidly hydrolyzed due to
enzymaticdegradation and therefore, more stable and potent analogs
havebeen developed. These analogs were synthesized by
replacingL-amino acids with their D-isomers and reducing the length
bykeeping only the peptide epitope responsible for the
biologicalactivity. The most widely known analogs of somatostatin
arecyclic peptides named octreotide
(d-Phe-c[Cys-Phe-d-Trp-Lys-Thr-Cys]-Thr-ol), lanreotide
(d-2Nal-c[Cys-Tyr-d-Trp-Lys-Val-Cys]-Thr-NH2) and vapreotide
(d-Phe-c[Cys-Tyr-d-Trp-Lys-Val-Cys]-Trp-NH2), which bind mainly to
the subtype 2 recep-tor (SSTR2) that is known to be the most
frequently overex-pressed SSTR [68]. There are several examples of
PDCsconsisting of the aforementioned somatostatin targetingpeptides
[67,69,70], as also other somatostatin peptide analogs,e.g.,
pentetreotide (DTPA-d-Phe-c[Cys-Phe-d-Trp-Lys-Thr-Cys]-Thr-ol)
[71].
Epidermal growth factor (EGF): Epidermal growth factor re-ceptor
(EGFR) is a transmembrane protein belonging to theErbB family of
receptor tyrosine kinases which consists of4 structurally-related
members: EGFR/HER1 (ErbB-1), HER2/neu (ErbB-2), HER3 (ErbB-3) and
HER4 (ErbB-4). Cohen andRita Levi-Montalcini shared the Nobel Prize
in Medicine in
1986 for discovering growth factors. EGFR is upregulated in
awide pool of cancer tissues and is able to enter cells usually
viaclathrin-mediated endocytosis [72]. Many peptides have
beendiscovered to bind the EGFR with high affinity and
selectivitythrough screening phage display libraries and have been
usedlike a viable approach for targeted drug delivery:
YHWYGYT-PQNVI [73], CMYIEALDKYAC [74], LTVSPWY [75],YWPSVTL
[76].
Angiopep-2: A peptide that has recently attracted attention is
a19-mer peptide named angiopep-2 (TFFYGGSRGKRNNFK-TEEY), due to its
ability to cross the blood-brain barrier (BBB).The BBB is formed by
the endothelial cells of the brain,restricting and controlling the
exchange of molecules betweenthe central nervous system and the
rest body. Angiopep-2 isable to cross the BBB via receptor-mediated
transcytosis afterbinding to the low-density lipoprotein
receptor-related protein-1(LRP-1), which is overexpressed in brain
cells [77]. Moreover,the two lysines available in its sequence
render angiopep-2 anappealing PDC candidate, with the aim to
smuggle therapeuticpayloads to brain malignancies [78,79].
Cyclic peptide variants have been developed for the RGDpeptide
motif, reported above. The most commonly used cyclicpeptide is iRGD
(CRGDKGPDC), a 9-amino acid cyclicpeptide, with tumor tissue
penetration activity [80]. iRGDinitially binds to αVβ3 and αVβ5
integrins that are overex-pressed in tumor endothelial cells.
Afterward, a proteolyticalcleavage takes place to reveal a cryptic
RXXK/R motif locatedat the C-terminus (CendR motif, C-End Rule),
which then bindsto neuropilin-1 (NRP-1), activating an endocytic
transport path-way responsible for the enhanced transport of
anti-cancer drugsinto tumors (Figure 3) [80].
In Table 1 are reported the most common peptides (linear
andcyclic) utilized in PDCs.
Selecting the proper cytotoxic agent togenerate the
PDCsAccording to the National Cancer Institute (cancer.gov),
thereare more than 250 FDA-approved anticancer drugs utilized
totreat malignancies at the moment. Among this large pool
ofcytotoxic drugs, an array of them has been utilized as
toxicwarheads in PDCs and five representative examples are
gemci-tabine, doxorubicin, daunorubicin, paclitaxel and
camptothecin(Figure 4). The main drawback of these original
anticanceragents is their uncontrolled toxicity which results in
severe sideeffects. Without the addition of a targeting moiety,
they bearlow capacity to discriminate cancerous from normal
cells.Moreover, the addition of a peptide as a targeting vehicle
canenhance the pharmacokinetic and therapeutic window of the
-
Beilstein J. Org. Chem. 2018, 14, 930–954.
936
Table 1: The most common peptides (linear and cyclic) utilized
for the formulation of PDCs used in cancer. Letters with bold color
stand for D-aminoacids.
peptide name peptide sequence targeted receptor reference
RGD R-G-D integrin αvβ3 [37,46,51,52]iRGD CRGDK/RGPD/EC integrin
αvβ3/αvβ5 [81]
octreotide SSTR2/5 [69]
D-Lys6-LHRH Glp-H-W-S-Y-K-L-R-P-G LHRH-R [18,19,61]angiopep-2
T-F-F-Y-G-G-S-R-G-K-R-N-N-F-K-T-E-E-Y LRP-1 [78,79]
GE11 Y-H-W-Y-G-Y-T-P-Q-N-V-I ErbB1 (EGFR) [73]
Figure 3: Binding and penetration mechanism of iRGD. The
iRGDpeptide is accumulated on the surface of αv integrin-expressing
endo-thelial and other cells in malignancies. The RGD motif is
responsiblefor binding to integrins. Afterward, the peptide is
cleaved by cell sur-face-associated protease(s) to eventually
expose the cryptic CendR el-ement, RXXK/R, at the C-terminus (red
dotted line). The CendR ele-ment then interferes with the binding
to neuropilin-1, resulting in tissueand cell penetration. The
tumor-penetrating peptide can be used todecorate a cargo (a simple
chemical moiety or a nanoparticle), but onlyin the case that the
cargo is attached to the N-terminus of the iRGDpeptide as the
disulfide bond is cleaved before the peptide is internal-ized
(black line). The figure was adopted from reference [81] (©
2009Elsevier Ltd.).
parent cytotoxic agent. Since different drugs may employ a
dif-ferent mechanistic approach to kill cells, the appropriate drug
isselected according to features characterizing the
targetedcancerous cells. For instance, daunorubicin and
doxorubicinpossess similar mechanisms of action [82], whereas
gemcitabi-ne [83], camptothecin [84] and paclitaxel [85] function
throughdifferent mechanisms.
The selected drug must comply with certain design principles
inorder to serve as an appealing candidate for PDCs. The
selecteddrug must be amenable to the linker chemistry. It must bear
an
intrinsic functional group for direct conjugation with
thepeptide/linker (Figure 4) or a functional group able to be
deriva-tized for further conjugation (i.e., click chemistry [86]).
In thelatter case, the site of derivatization has to be carefully
selectedso that the biological activity of the drug and the release
of theactive drug will not be perturbed. In case that the drug
bindsthrough recognition of a specific receptor, in silico
approacheshave to be recruited in order to rationally select the
location ofthe drug that will be chemically modified [18].
Furthermore, it must be sufficiently cytotoxic versus
theselected malignant tumor cells in order to eliminate them
andconsequently promote tumor regression. The selected drugshould
ideally possess low-nanomolar IC50 values for thetargeted malignant
tumor. A legitimate strategy to overcome alow drug potency problem
is by increasing the drug loading ofthe peptide-carrier. For
example, in the PDC ANG1005, 3 drugmolecules (paclitaxel) were
loaded on a single angiopep-2peptide which has completed phase II
clinical trials [87]. Never-theless, the concept of higher drug
loading is hard to be imple-mented, in contrast with single drug
loading that is usuallypreferred, mostly due to poor
physicochemical properties.
Below we analyze a set of drugs that have been tailored and
in-corporated in PDCs.
Gemcitabine (Gem): Gemcitabine (dFdC) is a nucleosideanalog of
deoxycytidine in which the hydrogen atoms on the2' carbon are
replaced by fluorine. It is sold under the brandname Gemzar by Eli
Lilly and Company and has been FDA ap-proved for the treatment of
various cancers including breast,ovarian, non-small cell lung and
pancreatic cancer. The maindrawbacks for its use are the high and
non-selective toxicity tonormal cells, the deactivation through
deamination in its inac-tive metabolite dFdU, the acquired
multidrug resistance (MDR)and its high hydrophilicity deterring its
prolonged drug releasefrom various vehicles [88], which therefore
reduces the effec-tive concentration of gemcitabine. It enters
cells throughnucleoside transporters hENTs (human equilibrative
nucleoside
-
Beilstein J. Org. Chem. 2018, 14, 930–954.
937
Figure 4: Representative examples of anticancer drugs utilized
for the construction of PDCs. The most usual conjugation sites are
marked with redcycles.
transporters) and hCNTs (human concentrative
nucleosidetransporters) and mostly through hENT1 (human
equilibrativenucleoside transporter 1) [89,90]. After
internalization, gemcita-bine is sequentially mono-, di- and
tri-phosphorylated by phos-phorylating kinases. Gemcitabine
diphosphate (dFdCDP) andgemcitabine triphosphate (dFdCTP) are the
active metaboliteswhich inhibit processes required for DNA
synthesis [91]. Theincorporation of dFdCTP into DNA during
polymerization,which causes DNA polymerases unable to proceed, is
the majormechanism by which gemcitabine causes cell death
(maskedtermination) [83]. Regarding the possible functional sites
ingemcitabine that can be used for the construction of PDCs areits
primary and secondary alcohols as also the amine (Figure 4).
Paclitaxel (PTX): Paclitaxel (PTX) is a member of the
taxanefamily and one of the most common anticancer agents
usedagainst a wide variety of tumors. It is sold under the brand
nameTaxol by Bristol-Myers Squibb Company and is FDA approvedfor
the treatment of breast cancer, ovarian cancer, non-small celllung
cancer and AIDS-related Kaposi's sarcoma. The maindisadvantages in
the utilization of paclitaxel are its high hydro-phobicity,
requiring suitable vehicles to effectively deliver it totumor
tissues, and the development of multidrug resistance due
to the P-glycoprotein-mediated efflux [85,92]. Paclitaxel
stabi-lizes microtubules by binding specifically to the
beta-tubulinsubunit, promoting mitotic halt and consequently cell
death[93]. The difference with other known drugs that act on
micro-tubules (vinca alkaloids) is that paclitaxel does not induce
thedisassembly of microtubules but boosts the polymerization
oftubulin [94]. Sites available in PTX for the formation of PDCsare
highlighted in Figure 4.
Anthracyclines: Anthracyclines are among the main anti-cancer
drugs that are applied in combinations with otherchemotherapeutic
agents. They are utilized against a variety ofcancers including
leukemias, lymphomas, breast, ovarian,bladder and lung.
Daunorubicin (Dau) was the first anthracy-cline discovered that was
extracted from Streptomycespeucetius, a species of actinobacteria,
at the beginning of the1960s. Shortly after, the isolation of
doxorubicin (Dox) from amutated Streptomyces strain was
accomplished. Anthracyclinesare consisted of a tetracyclin aglycon
part and a daunosaminesugar moiety. The difference between Dau and
Dox is ahydroxy group substituted at the C-14 carbon atom on Dox
pro-viding an extra conjugation site for ester linkage (Figure 6).
Themechanism of action of anthracyclines is based on their
interca-
-
Beilstein J. Org. Chem. 2018, 14, 930–954.
938
lation to DNA inhibiting the macromolecular
biosynthesis.Furthermore, they stabilize the topoisomerase II DNA
complexpreventing the transcription. They may also increase
quinonetype free radical production, however, this plays a role
rather intheir cytotoxic side effects. Daunorubicin is mainly used
in thetreatment of leukemia [95] while doxorubicin in the cure
ofother types of cancers (breast cancer, bladder cancer,
Kaposi'ssarcoma) in combination with other anti-cancer agents.
Camptothecin (CPT): Camptothecin is a cytotoxic
alkaloidcollected from extraction of the bark and stem of the
Chinesetree ‘Camptotheca acuminata’. It was first isolated and
charac-terized in 1966 by Wall et al. [96,97]. The main mechanism
ofaction involves binding to the reversible complex of
topoisom-erase I (topo I) and the 3′-phosphate group of the DNA
back-bone through hydrogen bonding, resulting in accumulation of
apersistent ternary complex (the cleavable complex). This
stabi-lized complex prevents the re-ligation step of DNA,
catalyzedby topo I, resulting in DNA damage and therefore cell
death(apoptosis). CPT is predominantly cytotoxic during the S
phasereplication of DNA because of the collision of the
replicationfork with the cleavable complex, converting the
single-strandbreaks into double-strand breaks and eventually
causing celldeath [98]. CPT can be conjugated to targeting elements
to en-hance its efficacy via its primary alcohol marked in Figure
4.Although CPT showed remarkable results during its phase Iclinical
trials against a variety of solid tumors, its low water-solubility
and stability led to the formulation of various newanalogs with the
same mechanism of action. The two mostprogressed analogs of CPT are
topotecan and irinotecan.Topotecan (hycamtin) has been approved by
the FDA for thetreatment of ovarian and cervical cancer, as also
small cell lungcarcinoma. Irinotecan (camptosar) has been approved
by theFDA for the treatment of metastatic carcinoma of the colon
orrectum, alone or in combination with fluorouracil
(5-FU).Camptothecin has been utilized as an anticancer agent
invarious PDC formulations, such as conjugation with thetargeting
peptides D-Lys6-LHRH [99], somatostatin [100] andc(RGDyK)
[101].
Linker design for PDCs: Principles andrepresentative
examplesAnother crucial aspect that should be considered during
thedesign of a PDC is the linker tethering the peptide and the
drug.The linker has to be carefully shaped so as not to perturb
thebinding affinity of the peptide to its receptor and the drug
effi-cacy. An inappropriate linker may impede the release of
thedrug from the PDC and therefore diminish its overall
thera-peutic potency. Linkers utilized in PDCs exist in different
cate-gories and vary on their length, stability, release
mechanism,functional groups, hydrophilicity/hydrophobicity etc.
This linker can be designed to bear an enzyme-hydrolyzableunit
(EHU) like a carboxylic ester or an amide bond, cleaved byesterases
and amidases, respectively. The most commonlyutilized linkers that
bear a carboxylic ester bond, as the enzyme-hydrolyzable unit, are
succinyl (derived from succinic acid) andglutaryl (derived from
glutaric acid). Concerning the utilizationof amide bond in the
linker as the unit tethering the drug and thepeptide, it can be
tailored to be cleaved based on the targetedtissue and/or type of
cancer where a specific protease is statisti-cally upregulated
(i.e., cathepsin B upregulated in variousmalignancies including
lung, brain, prostate and breast [102]).Also, during the design of
the PDC specific attention has to begiven on the selection of the
bonds that will be used in thelinker. Specifically, in several
currently available PDCs, at leasttwo different bonds are used: one
to connect the linker to thepeptide and the other to connect the
drug to the linker. Suchcases have to consider, during the design
process, the microen-vironment that the assembled PDC is to be
located, since differ-ent enzymes and/or the tumor microenvironment
might triggerthe improper release of the drug from the PDC, i.e.,
to end upwith the drug-carrying part of the linker.
Another class of linkers is the
stimuli-responsive/degradablelinkers, designed to achieve an
efficient release of the drug fromthe bioconjugate in the tumor
microenvironment. Such linkersare rationally designed to be cleaved
when they sense specificstimuli in the environment of cancerous
cells (slightly acidicpH, enhanced levels of reducing agents and/or
enzymes) orexternal stimuli (ultrasound, temperature, irradiation).
Specifi-cally, there are certain bonds like imine, oxime,
hydrazone,orthoester, acetal, vinyl ether and polyketal [103] that
areknown to undergo hydrolysis at acidic pH, while beingextremely
stable during blood circulation. Therefore, acid-labilebonds could
be hydrolyzed in the slightly acidic microenviron-ment and/or in
the acidic cellular compartments of cancer cellsand consequently
release the active drug. Additionally, disul-fide linkers are often
adopted in PDCs, since they are cleavedby reducing agents like
cysteine and glutathione, present in highconcentrations in
malignant cells.
Linkers bearing enzyme-hydrolyzable units (EHU) responsiveto
proteases are degradable peptide linkers that have
attractedsignificant interest due to the specificity of certain
enzymes andthere has been a dramatic escalation over in the past
years. Themost representative examples in this field are the
MMP-2/9(matrix metalloproteinases) and cathepsin B peptide
substrates.MMP-2/9 and cathepsin B are proteolytic enzymes present
atelevated levels in cancer cells known to participate in
humantumor invasion and metastasis. Cathepsin B is able to
recognizespecific peptide sequences like Val-Cit
(valine-citrulline) [104]and GFLG [105]. On the other hand,
GPLGIAGQ [106],
-
Beilstein J. Org. Chem. 2018, 14, 930–954.
939
Figure 5: Illustration of the drug release mechanism from the
self-immolative spacer PABC conjugated to a tumor homing peptide
via an enzyme-hydrolyzable unit. Red color = the self-immolative
spacer PABC; blue color = drug; green color = enzyme-hydrolyzable
unit (EHU); black color thetumor-homing peptide.
Table 2: Representative examples of biodegradable/responsive
linkers utilized for the formulation of PDCs in cancer.
linker drug release mechanism reference
succinyl action of esterases/amidases [19]glutaryl action of
esterases/amidases [67]PABC 1,6-elimination [109,110]oxime bond
hydrolysis in acidic pH [111]peptide GFLG action of cathepsin B
[105]peptide PLGLAG action of MMP-2/9 [107]
PLGLAG [107] and GPVGLIGK [108] are some commonpeptide
substrates for MMP-2 and MMP-9.
Another rapidly emerging category in PDC linkers that hasgained
much attention in the last years are the self-immolativeor
self-destructive spacers/linkers [109,110]. This type
oflinkers/spacers offers the capability to release the active
drugafter simultaneous cascade reactions, as shown in Figure
5.Para-amino benzyl alcohol (PABC; colored in red) is a
repre-sentative example that can be connected in the amino group
viaan amide bond to an enzyme-hydrolyzable unit (EHU; coloredin
green) and to a tumor-targeting element (i.e. tumor homingpeptide;
colored in black). The alcohol group at the oppositesite can be
connected via a carbonate ester/carbamate bond tothe cytotoxic
agent (colored in blue). The EHU is designed soas to be a substrate
for proteases overexpressed in the targetedtumor microenvironment
(i.e cathepsin B). Once EHU will berecognized by these enzymes it
is cleaved off resulting in theconsequent release of the active
drug through rapid cascadereactions (Figure 5).
The most representative examples of various types of linkersare
summarized in Table 2.
Representative examples of PDCs targetingcancer cellsIntegrating
the basic design principles in PDCs pinpointedabove, a list of
representative developed examples is analyzedbelow, so as to
provide a spherical perspective regardingpeptide–drug conjugation
chemistry.
Currently, there are two PDCs that have been developedutilizing
peptides as tumor targeting elements that selectivelybind to
specific receptors and small molecules as anticanceragents that
have reached phase III clinical trials (Table 3) forthe treatment
of various types of cancer. ClinicalTrials.gov havealso announced
the initiation of a clinical trial based onvarious PDCs consisted
of two novel peptides selectedafter phage display that target
murine A20 leukemiccells (ClinicalTrials.gov Identifier:
NCT02828774). Theseclinical trials will focus on chronic
lymphocytic leukemia(CLL).
Except these two PDCs, there are other types of PDCs that donot
consist of peptides as targeting moieties and small mole-cules as
drugs and have reached even up to phase III clinicaltrials. These
PDCs are summarized in Table 4.
Notably, there is only one PDC in the market designated
111In-DTPA-d-Phe1-octreotide, which is utilized for diagnostic
radi-ology in somatostatin receptor-positive tumors [118]. It
consti-tutes a complex of 111Indium bound to
diethylenetriaminopen-taacetic acid (DTPA), which is conjugated to
the targetingsomatostatin peptide [D-Phe1]-octreotide. Recently,
anothersimilar analog, designated
111In-DTPA-d-Phe-1-Asp0-d-Phe1-octreotide, has been evaluated and
presented enhanced tumoraccumulation in pancreatic tumor cells and
simultaneouslylower renal radioactivity [119].
Herein, we will analyze in depth the two PDCs in clinical
trialsconsisted of peptides and small molecules (Table 3), as
also
-
Beilstein J. Org. Chem. 2018, 14, 930–954.
940
Table 3: Peptide–drug conjugates consisting of peptides and
small molecules that have been used in clinical trials.
peptide cytotoxic agent linker drug release mechanism name
target CCTa reference
D-Lys6-LHRH Dox (SM)b glutaryl esterases/amidases AEZS-108
LHRH-R phase III [112]angiopep-2 PTX (SM) succinyl
esterases/amidases ANG1005 LRP-1 phase II [79]
aCCT= current clinical trials; bSM= small molecule.
Table 4: Various other types of peptide–drug conjugates in
clinical trials.
peptide cytotoxic agent linker drug releasemechanism
target name CCTa reference
CNGRCG hTNFα (Protein) – amidases CD13 receptor NGR015 phase III
[113]polyglutamic acid PTX (SM)b – esterases – CT2103 phase III
[114]LHRH CLIP71c (lytic peptide) – amidases LHRH-R EP-100 phase I
[115]DRDDS (spacer) DAVBLHd (SM)b 2-mercapto-
ethanolglutathione folate receptor EC145 phase III [116]
D-γ-E-γ-E-γ-E-E(masking moiety)
12ADTe-Asp – PSMA PSMA G-202 phase II [117]
aCCT = current clinical trials; bSM = small molecule; cCLIP71 =
KFAKFAKKFAKFAKKFAK; dDAVBLH = desacetyl vinblastine
hydrazide;e12ADT = 8-O-(12-aminododecanoyl)-8-O-debutanoyl
thapsigargin.
Figure 6: Structures of the PDCs named AN-152 and AN-207.
various other similar PDC formulations existed in the
currentliterature that have been evaluated in preclinical
models.
First, two widely-known peptide–drug conjugates namedAN-152
(AEZS-108) and AN-207 will be analyzed. Theseconjugates contain the
luteinizing hormone-releasing hormone(LHRH) as the
peptide-targeting module and doxorubicin(DOX) or its
daunosamine-modified derivative 2-pyrrolino-DOX as the cytotoxic
agent, respectively (Figure 6). Specifi-cally, Andrew V. Schally
and his group first synthesized thecorresponding analogs [120]
where they covalently coupled the
two drugs to the epsilon-amino group of the D-Lys side chain
ofthe peptide D-Lys6-LHRH.
Notably, both conjugates fully preserved the cytotoxic
activityof the parent drugs, DOX or 2-pyrrolino-DOX, respectively,
invitro and also retained the high binding affinity of their
peptidecarrier to receptors for LHRH on rat pituitary [120]. The
twoconjugates were subjected to stability tests and they showedslow
drug release in human serum in contrast with nude micethat
carboxylesterase enzymes are about 10 times higher
[121].Consequently, the two analogs were heavily evaluated in
in
-
Beilstein J. Org. Chem. 2018, 14, 930–954.
941
vivo models in nude mice bearing various types of cancer.
Micebearing OV-1063 (LHRH receptor positive) or UCI-107(LHRH
receptor negative) human epithelial ovarian cancerswere treated
with AN-152 or DOX with systematic intraperi-toneal administration.
The growth of UCI-107 cells was not in-hibited by AN-152 but
systemic administration of AN-152 inOV-1063 cells proved that
AN-152 is less toxic but inhibitstumor growth better than equimolar
doses of DOX [122]. Theseresults were confirmed in nude mice
bearing other ovarianhuman cancers (ES-2), where AN-207 caused up
to 59.5% inhi-bition in tumor growth [123]. Also, AN-207 and AN-152
weretested in female BDF mice bearing estrogen independent MXTmouse
mammary cancers, presenting stronger tumor inhibitoryeffects than
their respective cytotoxic radicals up to 93%, whileequimolar
quantities of their respective radicals were more toxic[124].
Moreover, PDC AN-207 was significantly more potent,regarding the
growth inhibition of hormone-dependent DunningR-3327-H prostate
cancers in rats, reaching up to 50% of theinitial tumor volume in
comparison with 2-pyrrolino-DOX.Shortly afterward, they tested the
two conjugates in membranesof human breast cancer cells: MCF-7
hormone-dependent andMDA-MB-231 hormone- independent [125]. They
proved thatthe specific analogs retained the high binding affinity
of theD-Lys6-LHRH carrier to the relevant receptors. Bothconjugates
displayed IC50 values in the low nanomolarconcentration range for
MCF-7 (13.7 ± 1.09 nM for AN-152and 6.08 ± 0.5 nM for AN-207) and
MDA-MB-231(5.60 ± 1.24 nM for AN-152 and 1.89 ± 0.4 nM for
AN-207)cells. AN-152 was tested regarding the inhibition of
tumorgrowth of subcutaneously (sc) implanted
androgen-dependentLNCaP and MDA-PCa-2b and androgen-independent
C4-2prostate cancers, xenografted into nude mice. The
resultsdemonstrated the stronger inhibition of AN-152 on the
tumorwith respect to the free DOX [126]. Similarly, in vivo
experi-ments were conducted regarding AN-207 in nude mice
bearingxenografts of MDA-PCa-2b prostate cancer cells, showing
iden-tical results like AN-152 [127]. Gründker et. al. evaluated
theantitumor effects of AN-152 in vivo in human LHRH-R-posi-tive
HEC-1B endometrial and NIH:OVCAR-3 ovarian cancers,and in the
LHRH-R-negative SK-OV-3 ovarian cancer cell linevia intravenous
injections [128]. The tumor volumes of HEC-1B and NIH:OVCAR-3
cancers were reduced significantly evenafter 1 week of treatment
with AN-152 while presenting notoxic side effects. Treatment with
DOX arrested tumor growthbut did not reduce tumor volume. The
growth of SK-OV-3cancers was not affected by AN-152. Based on the
presentedresults, it can be concluded that these two analogs
possesshigher antitumor activity but less toxicity with respect to
theparent drugs DOX and 2-pyrrolino-DOX and can be usedversus a
wide variety of ovarian, prostate, endometrial andbreast
tumors.
Notably, after the extensive evaluation of analog AN-152
inpreclinical models, starting from 2006 it has been tested inphase
I and phase II studies (AN-152 was renamed to AEZS-108 for the
clinical trials) of LHRH-R positive recurrentendometrial and
ovarian cancers. The phase I/II study in castra-tion-resistant
prostate cancer (CRPC) and chemotherapy refrac-tory bladder cancer
also showed promising results. Due to thepromising results from
phase II trials in endometrial cancer, amultinational phase III
clinical study is underway [112].
It is important to note that despite the fact that analog
AN-207presented a better biological profile, evident in all the
preclin-ical models, its further development fell short due to
chemicaland plasma instability.
According to ClinicalTrials.gov, during phase I analog AEZS-108
was tested in 17 women with epithelial cancer of the
ovary,endometrium or breast and for which standard treatment
couldnot be used or was no longer effective. The results
showedpromising tolerance from the patients with fewer side
effectsthan the commonly applied drugs. Moreover, AEZS-108
wasevaluated in phase I clinical trial on patients with
castration-and taxane-resistant prostate cancer and the results
proved thatAEZS-108 possesses a sufficient safety profile and
efficacy. Itsucceeded in lowering the PSA levels in some patients
withprostate cancer and it became evident that the internalization
ofAEZS-108 in prostate cancer circulating tumor cells (CTCs)may be
a viable pharmacodynamic marker [129].
These promising results led to phase II clinical trials to
patientswith castration- and taxane-resistant prostate cancer and
theirdisease showed progression after taxane-based
chemotherapy.AEZS-108 showed significant activity in these patients
whowere pretreated with taxanes and maintained an acceptablesafety
profile [130].
Last, phase II clinical trials were conducted in
collaborationwith the German Gynecological Oncology Group (AGO)
and3 other centers from Bulgaria on 43 women. The patients
hadplatinum-resistant advanced ovarian cancer, FIGO
(FédérationInternationale de Gynécologie et d'Obstétrique) III or
IV orrecurrent endometrial cancer (EC) and LHRH
receptor-positivetumor status. The treatment with AEZS-108 had
significant ac-tivity and low toxicity in these women
[131,132].
Based on the fact that the previously described analog
AN-207showed superior in vitro and in vivo results compared
toAN-152 but lacked stability, Andrew V. Schally and his
groupturned their focus on its building block 2-pyrrolino-DOX
andtried to construct new PDCs using other peptides. Therefore,they
synthesized a new analog, designated AN-238, consisting
-
Beilstein J. Org. Chem. 2018, 14, 930–954.
942
Figure 7: Structure of the PDC named AN-238.
of the octapeptide RC-121 linked through the α-amino group ofits
N-terminal D-Phe moiety and a glutaric acid spacer to the14-OH
group of 2-pyrrolino-DOX (Figure 7). The octapeptideRC-121 was
utilized due to its high binding affinity to thesomatostatin
receptor (SST-R) [133].
The anti-cancer activity was first evaluated in various
rat/humancancer lines xenografted into nude mice with breast
humantumors (MDA-MB-238, MCF-7, MX-1) and prostate rat/humantumors
(Dunning AT-1, PC-3). All cell lines showed a greatresponse to the
treatment with AN-238 with high inhibition ofthe tumor, while 5 of
10 mice with MX-1 tumor were totallycured [61]. The cytotoxic
profile of this analog was similarlyevaluated in additional cancer
cell lines xenografted into nudemice including prostate, renal,
mammary, ovarian, gastric,colorectal and pancreatic [134]. Various
types of renal,colorectal, pancreatic and gastric cancers showed a
majorresponse to the treatment with more than 70% inhibition
whileall the other types showed a good response to the treatment
withan average of 60% inhibition. AN-238 was also evaluated
inU87-MG brain cancer cells with good response, inducing82% growth
inhibition of subcutaneous tumors [134]. There-fore, AN-238 has
been proved to be a promising candidate for alarge number of
tumors, being able to suppress the growth ofthese tumors and their
metastases. Last, Engel et. al. showedthat AN-238 inhibits tumor
growth in human experimentalendometrial carcinomas which express
SST receptors, regard-less of the expression levels of multidrug
resistance proteinMDR-1 [135]. The analog AN-238 is still pending
for clinicaltrials.
An interesting example of a PDC able to cross the
blood-brainbarrier (BBB), is ANG1005 [136], composed of three
mole-cules of paclitaxel linked by a cleavable succinyl ester
linkageto the angiopep-2 peptide (Figure 8).
BBB is formed by the brain capillary endothelium with verylow
permeability as it excludes about 100% of the large mole-cules and
about 98% of the small molecules attempting to pass
to the brain [137]. Being mandatory to surpass the BBB in
orderto deliver pharmaceuticals to the brain, scientists have
strug-gled to discover either novel small molecules able to cross
itthrough various mechanisms [138] or novel techniques able
todisrupt its dense structure like ultrasound-mediated drugdelivery
[139,140]. The design principles on the synthesis of thespecific
conjugate, ANG1005, were the following: the peptideangiopep-2 is
able to cross the BBB via receptor-mediated tran-scytosis after
binding to LRP-1 and consequently it is oftenused as drug delivery
vehicle, while paclitaxel bears cytotoxici-ty against glioblastoma.
It has been shown that the brain uptakeof ANG1005 was 4.5-fold
higher compared to paclitaxel andthe cytotoxicity remained higher
in all cancer cell lines tested(glioblastoma U87 MG, U118, U251;
lung carcinoma A549,NCI-H460, Calu-3; ovarian carcinoma SK-OV-3).
It has beenalso proved that human tumor xenografts were inhibited
morewith ANG1005 than paclitaxel. Finally, mice with
intracerebralimplantation of U87 MG glioblastoma cells or NCI-H460
lungcarcinoma cells exhibited increased survival rates afterANG1005
administration.
Because of these promising results, ANG1005 progressed tophase I
clinical trials in 2007 in 63 patients with recurrent orprogressive
malignant glioma. It was found that ANG1005delivers paclitaxel
across the BBB and achieves therapeuticconcentrations in the tumor
site. It became evident that thisPDC possessed similar toxicity to
paclitaxel as also enhancedactivity in recurrent glioma [141].
Phase II clinical trials werethen initiated on patients with
recurrent high-grade glioma andon breast cancer patients with
recurrent brain metastases. Theresults have not been published yet
but it has already beenstated that very promising results were
collected and phase IIIclinical trials will start shortly. Based on
the overall progress ofANG1005, other similar molecules have been
synthesized andstudied in preclinical models [142].
The group of Prof. G. Mező has achieved a great progression
inthe field of PDCs the last years working mostly on
GnRH-III(Glp-His-Trp-Ser-His-Asp-Trp-Lys-Pro-Gly-NH2), which
was
-
Beilstein J. Org. Chem. 2018, 14, 930–954.
943
Figure 8: Chemical structure and synthetic scheme for the PDC
ANG1005. (A) ANG1005 is composed of three molecules of paclitaxel
linked by acleavable succinyl ester linkage to the angiopep-2
peptide. (B) Schematic representation of ANG1005 synthesis steps.
Paclitaxel was first reactedwith succinic anhydride and then
activated with N-hydroxysuccinimide to form
2′-succinyl-NHS-paclitaxel in two steps. In the conjugation step
(step 3),amines of the angiopep-2 peptide react with
2′-succinyl-NHS-paclitaxel. The scheme was modified according to
Br. J. Pharmacol. 2008, 155, 185–197[136].
exploited as a tumor homing device for drug targeting 10
yearsago [111]. The aim of this was to apply a peptide hormone
withlower endocrine effect than GnRH-I that might be useful
espe-cially for hormone-independent tumors like colon cancer
[143].In addition, GnRH-III has Lys at position 8 of the sequence
pro-viding a conjugation site without inducing perturbation in
thereceptor recognition. In the first conjugates, daunorubicin
(Dau)was attached to the lysine side chain via oxime linkage
throughan aminooxyacetyl (Aoa) moiety
(Glp-His-Trp-Ser-His-Asp-Trp-Lys(Dau=Aoa)-Pro-Gly-NH2). The oxime
bond, de-
veloped between the aminooxyacetyl function and the
carbonylgroup of C-13 on Dau is stable under physiological
conditionsand prevents the early drug release in contrast to the
ester bond(Figure 9).
Thus, no free drug release can be detected from such type
ofconjugates before reaching their targets. However, oxime bondis
also stable in lysosomes where the conjugates decomposedafter
receptor-mediated endocytosis. Among the fragmentsarose during
lysosomal degradation, H-Lys(Dau=Aoa)-OH was
-
Beilstein J. Org. Chem. 2018, 14, 930–954.
944
Figure 9: Structure of oxime linked Dau–GnRH-III conjugate with
or without cathepsin B labile spacer and their metabolite released
in lysosomalhomogenate [144].
observed as the smallest Dau containing metabolite
[144].Therefore, the DNA binding propensity of this metabolite
wasalso examined and it was found that although it is efficient
itpresented lower binding capacity with respect to the
parentdrug.
The in vitro antitumor activity of the above-mentioned
conju-gate was studied on MCF-7 human breast and HT-29 humancolon
adenocarcinoma cells [144]. The IC50 values showed twoorders of
magnitude lower effect compared to the free Dau.Thus, a systematic
comparative study of various anthracycline-GnRH conjugates was
conducted in order to conduct their com-plete evaluation as
potential targeted cancer chemotherapeutics.The influence of
different: (i) anthracycline drugs, (ii) linkersamong the
tumor-homing peptide moiety and the drug, and (iii)tumor-homing
peptides (e.g., GnRH-III and D-Lys6-GnRH-I)was examined regarding
their in vitro cellular uptake, drugrelease and cytostatic effect
[145]. Doxorubicin (Dox) wascoupled to both GnRH-III and
D-Lys6-GnRH-I through aglutaric acid linker via ester bond. AN-152,
the GnRH-I basedPDC (see above), served as a control. No
significant differ-ences in cellular uptake and cytostatic effect
were observed be-tween the two PDCs. Recently, it was also
indicated that thecellular uptake of carboxyfluorescein-labeled
GnRH-I, GnRH-IIand GnRH-III conjugates might be influenced not only
by thetargeting moiety, but also by the type of cancer cells
[146].
However, no significant differences could be observedregarding
the cellular uptake of the three GnRH conjugates byMCF-7 and HT-29
cells. It is worth mentioning that the highestwater solubility was
detected for the GnRH-III conjugate. Theester bond can be cleaved
by esterases not only in cancer cells,but also in human plasma
during blood circulation. The earlydrug release in the bloodstream
may cause unwanted sideeffects. Furthermore, O–N acyl shift was
detected both duringthe synthesis and the storage of ester-linked
doxorubicin conju-gate, resulting to an inactive compound where the
tumor-homing peptide acylated the amine of the daunosamine
sugarmoiety. This was found through the mass spectrometric
(MS)fragmentation profile of the PDC [146].
In a different PDC, Dau was linked to GnRH-III through
ahydrazone bond or by incorporation of a self-immolative spac-er
[145]. The hydrazone linkage was formed similarly to theoxime bond
on C-13 atom of Dau but it allows the effectivedrug release under
slightly acidic conditions in lysosomes.
Thep-aminobenzylalcohol-based self-immolative spacer, combinedwith
the dipeptide Lys-Phe (cathepsin-B lysosomal enzymecleavable
spacer), was connected to the amino functional groupof daunosamine
moiety. The last construct also provided thefree drug release. Both
conjugates illustrated similar cytostaticeffects and cellular
uptake as the conjugates with ester bonds.All these conjugates
showed IC50 values in the range of
-
Beilstein J. Org. Chem. 2018, 14, 930–954.
945
0.2–0.5 μM on MCF-7 cells while 1–3 μM on HT-29 cells. Thefree
Dau or Dox had higher in vitro cytostatic effect than
theconjugates, especially on HT-29 cells. Nevertheless, the
synthe-sis of these conjugates was not so efficient and their
chemical,biological and long-term shelf-stability of these PDCs
were notso sufficient for drug development.
In another construct, daunorubicin and doxorubicin were
at-tached to the ε-amino group of Lys of GnRH-III through
oximelinkage [111,145]. The conjugation of the drug and
theaminooxyacetylated tumor homing peptide was almost quantita-tive
under slightly acidic conditions. Interestingly, the conju-gate
with Dox illustrated much lower antitumor effect than theDau
conjugate. The oxime-linked Dau-GnRH-III conjugate(non-cleavable
linker) had one order of magnitude lower anti-tumor activity than
the conjugates with the cleavable linkers.The cellular uptake of
the oxime-linked conjugates was lower,too, but this effect might
come from the different fluorescentproperties of the free Dau and
the peptide/metabolite-linkedDau. Because of the high synthetic
yield and stability of oxime-linked conjugates, it can be suggested
that such conjugatesmight be good candidates for the development of
targeted tumortherapeutics. Therefore, efforts were made to develop
furtherconjugates with higher antitumor activity.
To achieve higher antitumor activity, the sequence of thepeptide
GnRH-III was modified. Previous studies indicated thatonly a few
changes are acceptable without significant loss of
theanti-proliferative effect of the hormone peptide.
Interestingly,Ser at position 4 could be replaced by Lys or
acetylated Lys[147]. It is worth mentioning that the Ser in GnRH
agonist andantagonist analogs are rarely modified [148]. The
incorporationof Lys or Lys(Ac) in position 4 of GnRH-III increased
the anti-tumor activity of the conjugate GnRH-III(Dau=Aoa).
However,in the case of [4Lys]-GnRH-III(Dau=Aoa) enzyme stability
ofthe conjugate was decreased while [4Lys(Ac)]-GnRH-III(Dau=Aoa)
showed higher stability [149]. When the acetylgroup was exchanged
to other short-chain fatty acids (SCFAs)the enzyme stability was
enhanced by the length of hydro-carbon chain of SCFAs [150].
According to the cellular uptakeand cytostatic in vitro studies,
the optimal compound was thebutyric acid containing
[4Lys(Bu)]-GnRH-III(Dau=Aoa) conju-gate that almost reached the in
vitro biological effects of theconjugates with a cleavable linker.
This conjugate showed sig-nificant tumor growth inhibition in vivo,
not only on subcuta-neous implanted but also on orthotopically
developed HT-29colon cancer-bearing mice [151]. The PDC in the
applied dose(15 mg Dau content/kg body weight) showed similar or
higherantitumor activity than the free Dau at a maximal tolerated
dose(MTD) without significant toxic side effects on organs.
Incontrast to the conjugate, Dau presented toxicity on the
liver
causing worse condition and higher mortality during the
treat-ment.
In addition, the incorporation of Lys at position 4 provided
anew conjugation site. Therefore, Dau or methotrexate (MTX)were
attached to the ε-amino group of 4Lys resulting in conju-gates with
two identical ([4Lys(Dau=Aoa), 8Lys(Dau=Aoa)]-GnRH-III), or
different drug molecules ([4Lys(MTX),8Lys(Dau=Aoa)]-GnRH-III)
[152,153]. Some improvement inthe cytostatic effect could be
detected compared with the conju-gates containing only one drug
molecule, but they were notbetter than the conjugate with butyric
acid. This observation ledto retain the Lys(Bu) at position 4 and
the two Dau moleculeswere attached to the amino groups of an
additional Lysthrough the enzyme labile GFLG spacer coupled to
8Lys(Figure 10).
The resulted PDC presented a reduced aqueous solubility, thusan
oligoethylene glycol linker was inserted between the spacerand the
tumor-homing peptide [154]. This PDC showed the bestin vitro
cytostatic effects among the oxime-linked Dau-contain-ing
conjugates, but the improvement of the synthetic process tolead to
higher amounts of this PDC is required to proceed for invivo
studies. Thus, it can be concluded that oxime linkedDau–homing
peptide conjugates could be good candidates fortargeted tumor
therapy.
Furthermore, the tumor homing peptide D-Lys6-GnRH-I hasbeen
exploited by our group to selectively deliver the anti-cancer agent
gemcitabine to the tumor site. We, therefore, de-signed and
synthesized four different bioconjugates consistingof D-Lys6-GnRH-I
and the anticancer agent gemcitabine(named GSG, GSG2, 3G, 3G2)
through different conjugationsites (the primary and secondary
alcohol groups of gemcitabine)and using linkers of different
lengths (succinyl and glutaryl) asshown in Figure 11.
In order to evaluate whether the tethering of the cytotoxic
agentto the D-Lys6-GnRH-I peptide induces any perturbation on
thelocal microenvironment of the peptide that is responsible for
re-ceptor recognition, we utilized 1H 1H 2D-TOCSY NMR [19].Upon
superimposing the relevant spectra of the different PDCson the
relevant spectrum of the native hormone we found thatthese PDCs
didn’t alter the microenvironment of D-Lys6-GnRH-I allowing to
suggest that they will not influence thebinding affinity of the
targeting peptide unit of these PDC to theGnRH-R. This was further
validated since the new conjugateswere found to possess higher
binding affinity with respect to theparent peptide, with IC50
ranging even up to 1.9 nM for theconjugate 3G. The conjugates were
evaluated regarding theirantiproliferative effect on prostate
cancer cells (DU145 and
-
Beilstein J. Org. Chem. 2018, 14, 930–954.
946
Figure 10: Synthesis of the most effective GnRH-III–Dau
conjugate with two drug molecules [153].
PC-3) and the PDC GSG showed IC50 values similar to gemci-tabine
GSG that possessed the highest antiproliferative effectwas utilized
for further pharmacokinetic studies in mice. Thesepinpointed that
GSG is able to release a high amount of gemci-tabine (averaging 500
ng/mL) for a period of over ≈250 min,while administered free
gemcitabine was consumed in less than100 min. At the same time, the
levels of the inactive metaboliteof gemcitabine (dFdU) were
maintained at very low levels forthe GSG conjugate in contrast to
the direct administration offree gemcitabine. Finally, when
injected into mice withxenografted tumors, GSG inhibited the tumor
growth moreeffectively than gemcitabine when using equimolar
quantities.Therefore, GSG could pave the way for the construction
ofother similar bioconjugates in order to effectively enhance
theconcentration of the cytotoxic drug in the tumor cells
andinhibit their uncontrolled growth.
OuWe have also designed and synthesized a PDC containingthe
cytotoxic agent sunitinib and the D-Lys6-GnRH peptide-
targeting-unit tethered through a succinyl linker [7]. Sunitinib
isa small orally administrated drug that inhibits the
phosphoryla-tion of several receptor tyrosine kinases (RTKs). It
was ap-proved by the FDA in 2006 for the treatment of renal cell
carci-noma (RCC) and imatinib-resistant gastrointestinal
stromaltumor (GIST). Though, sunitinib has proved to cause
severeside effects like cardiac and coronary microvascular
dysfunc-tion [155]. Therefore, these data rendered sunitinib as
anappealing candidate for targeted therapy using a PDC.
Native sunitinib (Figure 12A) does lack functional groups
thatcould be exploited for conjugation to the peptide-targeting
unit,thus, a novel analog had to be constructed (SAN1, Figure
12B).This was constructed based on in silico studies and
modifyingproperly the drug scaffold so as not to perturb the drug
bindingto the targeted receptors. In silico, in vitro and
pharmacokineticevaluation of the synthesized SAN1 were conducted
and com-pared with native sunitinib. The results indicated that
SAN1exhibited similar properties and thus could serve as an
alterna-
-
Beilstein J. Org. Chem. 2018, 14, 930–954.
947
Figure 11: Structures of the four different PDCs of
D-Lys6-GnRH-I and gemcitabine (GSG, GSG2, 3G, 3G2) [19].
Figure 12: Structures of (A) native sunitinib; (B) SAN1 analog
of sunitinib and (C) assembled PDC named SAN1GSC [18].
tive to the parent drug for the formulation of the final PDC.
Ad-ditionally, SAN1 was further explored in in vivo models:
micexenografted with a castration-resistant CaP (CRPC) cell
line
were subjected to treatment based on SAN1 and sunitinib.
Micewere dosed daily via intraperitoneal injection and the
resultsunveiled the potency of SAN1, which showed to inhibit
the
-
Beilstein J. Org. Chem. 2018, 14, 930–954.
948
tumor growth in a similar way like native sunitinib. In
theinstalled hydroxy group in the core of SAN1, a succinyl
linkerwas conjugated that was then connected to the free amine
groupof Lys8 of D-Lys6-GnRH to form the PDC named SAN1GSC(Figure
12C).
SAN1GSC was evaluated in vitro and then in vivo, in
micexenografted with the CRPC model, showing similar bioactivityas
SAN1. The most promising results arose from the
measuredconcentration of SAN1 in the blood circulation and in the
tumorsite. The levels of free SAN1 released from SAN1GSC were
4times higher inside the malignant cells with respect to SAN1levels
from the unconjugated SAN1. It is worth mentioning thatin the frame
of our construct, cardiotoxic and hematotoxiceffects in treated
mice were minimal and elevations of bloodpressure that contribute
to cardiac dysfunction were absent [18].
Problems and solutions during synthesis ofPDCsAlthough the
synthesis of PDCs is usually a rapid and facileprocedure, various
synthetic problems may arise. The mostcommon ones appear during
peptide synthesis and might referto low aqueous solubility and/or
difficulty to synthesize. Insolu-bility issues can be overcome by
altering the C-/N-terminusand/or substituting specific residues.
Difficulties in the synthe-sis can be handled by decreasing the
number of hydrophobicresidues and/or shortening the sequence.
Similar synthetic prob-lems have been encountered during peptide
synthesis the lastdecades and have been fully addressed.
During the conjugation of 5’-O-gemcitabine hemisuccinate tothe
D-Lys6-GnRH peptide towards the synthesis of the PDCGSG (presented
in Figure 10) we recently unveiled the forma-tion of a previously
unnoticed side product in addition to thedesired product [156].
Specifically, we found that when guani-dinium salts are utilized in
peptide coupling conditions, auronium derivative can be installed
on specific amino acid scaf-folds, beside to the formation of the
expected amide bonds. Thisside product was persistent even after
HPLC purification andwas also apparent in the recorded mass
spectrum of GSG as asecond peak, besides the expected product,
bearing the mass ofthe expected PDC plus 100 amu, leading to the
reduction of theoverall yield below 10%. Specifically, the
guanidinium/uroniumcoupling reagent (HATU) was utilized for the
formation of theamide bond between D-Lys6 of the peptide carrier
and thecarboxylic acid of the succinate linker connected to
gemcitabi-ne to synthesize the PDC GSG (Figure 13). We
hypothesizedthat the side product was originated from the coupling
reagent(HATU) and after conducting template reactions with
everyamino acid present in the sequence of D-Lys6-GnRH
withFmoc-Ser(t-Bu)-OH in the presence of HATU or HBTU and
DIPEA, it became evident that the aminium moiety of HATU/HBTU
could be installed either on the amino (–NH2) group ofLys or on the
phenol (–OH) group of Tyr [156].
Our findings were further verified by reacting other
tumor-homing peptides like D-Lys6-GnRH and Fmoc-HER2-BP1(LTVSPWY, a
heptapeptide known for its activity againsterbB2) with HATU/DIPEA
and characterizing the final prod-ucts by ESIMS and 1H NMR
spectroscopy where the sameaminium side product was also recorded.
Interestingly, whenthe dipeptide Fmoc-Cys-Tyr-NH2 was reacted with
HATU, wefound that the side product could be installed both on the
phenolgroup of Tyr as also on the sulfhydryl group of Cys.
Therefore,we tested these reaction conditions on the peptide
C1B5141–151subdomain peptide (RCVRSVPSLCG) of protein kinase C(PKC)
γ isozymes, which possess 2 cysteines (but no tyrosineor lysine or
free N-terminus amine) and bears anticancer prop-erties [157].
Again, a side product with two aminium moietieson the two cysteines
was formed and characterized withESIMS. This observation was also
applied to a simple phenolwhere again a side product was recorded
pinpointing the broadimpact of our findings beyond traditional
peptide chemistry. Wethus revealed the formation of a previously
unnoticed side prod-uct during the synthesis of PDCs, derived from
guanidinium/uronium peptide coupling reagents that occurs on
phenols, pri-mary amines and sulfhydryl groups (Figure 14).
Along these lines, we suggested a mechanism that this
side-product formation is taking place, directly after the
formation ofthe amide bond that occurs from structure (II) to
structure (III),as shown in Figure 15.
We discovered that the side product, which is difficult to
beseparated from the expected PDC and therefore results inreduced
synthetic yield, could be avoided by using 1 equiv ofHATU/HBTU,
instead of the classical and established proto-cols using 1.5 equiv
[156]. Taking into account that uronium/guanidinium coupling
reagents are among the most expensiveones, using the specified
conditions (equimolar quantity) mayalso reduce the total cost of
the synthesis.
ConclusionCurrently used chemotherapeutics are in their majority
highlytoxic, causing severe side effects. Thus, with the aim to
en-hance their narrow therapeutic index, a wide variety of
strate-gies have been explored. Selective drug delivery via
specialcarriers represents a viable approach to deal with tumors
withhigher efficacy while using lower doses of the anticancer
agent.Specifically, peptide–drug conjugates (PDCs) operate as
potentdrug delivery carriers and thus have attracted
considerableattention over the last decades. The simplicity,
versatility and
-
Beilstein J. Org. Chem. 2018, 14, 930–954.
949
Figure 13: Synthetic scheme for the formation of GSG and the
unexpected side product [156].
Figure 14: Illustration of uncharted guanidinium peptide
coupling reagent side reactions during PDCs synthesis [156].
-
Beilstein J. Org. Chem. 2018, 14, 930–954.
950
Figure 15: Putative mechanism for the formation of the uronium
side product [156].
the relatively low cost for the construction of PDCs
haverendered them appealing candidates. In the present review,basic
and integral knowledge has been accumulated towards thePDCs design
through examining every module required toassemble the fully
decorated PDC: the peptide, the cytotoxicagent and the linker. We
highlighted the overall progress of thisfield through selective
analysis of noteworthy examples in theliterature, as also possible
synthetic problems that may arise andtheir solutions. Based on the
fact that several PDCs have beenselected for clinical trials and
presented tumor inhibition withminimum side effects, this field
needs to be further refined andexplored. Through this review, we
made efforts to provide aninfluential impetus for the construction
of new peptide–drugconjugates, which could eventually transform
undesired toxicdrugs to highly potent formulations for the
effective treatmentof cancer.
AcknowledgementsThis work was co-financed by the European Union
(EuropeanSocial Fund ESF) and Greek national funds through the
Opera-tional Program “Education and Lifelong Learning” of
theNational Strategic Reference Framework (NSRF) - ResearchFunding
Program: ARISTEIA II [grant number: 5199]. Theauthors would also
like to thank National Research, Develop-ment and Innovation Office
(NKFIH K119552), Hungary.
ORCID® iDsEirinaios I. Vrettos -
https://orcid.org/0000-0002-4801-1900Gábor Mező -
https://orcid.org/0000-0002-7618-7954
References1. Organization, W. H.. The global burden of disease,
2004 update ed.;
World Health Organization: Geneva, 2008.2. Jemal, A.; Bray, F.;
Center, M. M.; Ferlay, J.; Ward, E.; Forman, D.
Ca-Cancer J. Clin. 2011, 61, 69–90. doi:10.3322/caac.201073.
Yan, L.; Rosen, N.; Arteaga, C. Chin. J. Cancer 2011, 30, 1–4.
doi:10.5732/cjc.010.105534. Mullard, A. Nat. Rev. Drug Discovery
2018, 17, 81–85.
doi:10.1038/nrd.2018.45. Aggarwal, S. Nat. Rev. Drug Discovery
2010, 9, 427–428.
doi:10.1038/nrd31866. Szakács, G.; Paterson, J. K.; Ludwig, J.
A.; Booth-Genthe, C.;
Gottesman, M. M. Nat. Rev. Drug Discovery 2006, 5,
219–234.doi:10.1038/nrd1984
7. Undevia, S. D.; Gomez-Abuin, G.; Ratain, M. J. Nat. Rev.
Cancer2005, 5, 447–458. doi:10.1038/nrc1629
8. de Sousa Cavalcante, L.; Monteiro, G. Eur. J. Pharmacol.
2014, 741,8–16. doi:10.1016/j.ejphar.2014.07.041
9. Bocci, G.; Kerbel, R. S. Nat. Rev. Clin. Oncol. 2016, 13,
659–673.doi:10.1038/nrclinonc.2016.64
10. Ferlay, J.; Steliarova-Foucher, E.; Lortet-Tieulent, J.;
Rosso, S.;Coebergh, J. W. W.; Comber, H.; Forman, D.; Bray, F. Eur.
J. Cancer2013, 49, 1374–1403. doi:10.1016/j.ejca.2012.12.027
https://orcid.org/0000-0002-4801-1900https://orcid.org/0000-0002-7618-7954https://doi.org/10.3322%2Fcaac.20107https://doi.org/10.5732%2Fcjc.010.10553https://doi.org/10.1038%2Fnrd.2018.4https://doi.org/10.1038%2Fnrd3186https://doi.org/10.1038%2Fnrd1984https://doi.org/10.1038%2Fnrc1629https://doi.org/10.1016%2Fj.ejphar.2014.07.041https://doi.org/10.1038%2Fnrclinonc.2016.64https://doi.org/10.1016%2Fj.ejca.2012.12.027
-
Beilstein J. Org. Chem. 2018, 14, 930–954.
951
11. Kola, I.; Landis, J. Nat. Rev. Drug Discovery 2004, 3,
711–716.doi:10.1038/nrd1470
12. Vander Heiden, M. G. Nat. Rev. Drug Discovery 2011, 10,
671–684.doi:10.1038/nrd3504
13. Galluzzi, L.; Kepp, O.; Vander Heiden, M. G.; Kroemer,
G.Nat. Rev. Drug Discovery 2013, 12, 829–846.
doi:10.1038/nrd4145
14. Bhat, M.; Robichaud, N.; Hulea, L.; Sonenberg, N.;
Pelletier, J.;Topisirovic, I. Nat. Rev. Drug Discovery 2015, 14,
261–278.doi:10.1038/nrd4505
15. Pfister, S. X.; Ashworth, A. Nat. Rev. Drug Discovery 2017,
16,241–263. doi:10.1038/nrd.2016.256
16. Tai, W.; Mahato, R.; Cheng, K. J. Controlled Release 2010,
146,264–275. doi:10.1016/j.jconrel.2010.04.009
17. Zwicke, G. L.; Mansoori, G. A.; Jeffery, C. J. Nano Rev.
2012, 3,No. 18496. doi:10.3402/nano.v3i0.18496
18. Argyros, O.; Karampelas, T.; Asvos, X.; Varela, A.; Sayyad,
N.;Papakyriakou, A.; Davos, C. H.; Tzakos, A. G.; Fokas,
D.;Tamvakopoulos, C. Cancer Res. 2016, 76,
1181–1192.doi:10.1158/0008-5472.CAN-15-2138
19. Karampelas, T.; Argyros, O.; Sayyad, N.; Spyridaki, K.;
Pappas, C.;Morgan, K.; Kolios, G.; Millar, R. P.; Liapakis, G.;
Tzakos, A. G.;Fokas, D.; Tamvakopoulos, C. Bioconjugate Chem. 2014,
25,813–823. doi:10.1021/bc500081g
20. Kellici, T. F.; Chatziathanasiadou, M. V.; Lee, M.-S.;
Sayyad, N.;Geromichalou, E. G.; Vrettos, E. I.; Tsiailanis, A. D.;
Chi, S.-W.;Geromichalos, G. D.; Mavromoustakos, T.; Tzakos, A.
G.Org. Biomol. Chem. 2017, 15, 7956–7976.
doi:10.1039/C7OB02045G
21. DeBerardinis, R. J.; Chandel, N. S. Sci. Adv. 2016, 2,
e1600200.doi:10.1126/sciadv.1600200
22. Trachootham, D.; Alexandre, J.; Huang, P. Nat. Rev. Drug
Discovery2009, 8, 579–591. doi:10.1038/nrd2803
23. Kato, Y.; Ozawa, S.; Miyamoto, C.; Maehata, Y.; Suzuki,
A.;Maeda, T.; Baba, Y. Cancer Cell Int. 2013, 13,
89.doi:10.1186/1475-2867-13-89
24. Singh, R.; Lillard, J. W., Jr. Exp. Mol. Pathol. 2009, 86,
215–223.doi:10.1016/j.yexmp.2008.12.004
25. Karakurt, S.; Kellici, T. F.; Mavromoustakos, T.; Tzakos, A.
G.;Yilmaz, M. Curr. Org. Chem. 2016, 20,
1043–1057.doi:10.2174/1385272820666151211192249
26. Kellici, T. F.; Chatziathanasiadou, M. V.; Diamantis,
D.;Chatzikonstantinou, A. V.; Andreadelis, I.; Christodoulou,
E.;Valsami, G.; Mavromoustakos, T.; Tzakos, A. G. Int. J. Pharm.
2016,511, 303–311. doi:10.1016/j.ijpharm.2016.07.008
27. Tsume, Y.; Incecayir, T.; Song, X.; Hilfinger, J. M.;
Amidon, G. L.Eur. J. Pharm. Biopharm. 2014, 86,
514–523.doi:10.1016/j.ejpb.2013.12.009
28. Apostolopoulos, V.; Pietersz, G. A.; Tsibanis, A.;
Tsikkinis, A.;Drakaki, H.; Loveland, B. E.; Piddlesden, S. J.;
Plebanski, M.;Pouniotis, D. S.; Alexis, M. N.; McKenzie, I. F.;
Vassilaros, S.Breast Cancer Res. 2006, 8, R27.
doi:10.1186/bcr1505
29. Tang, C. K.; Lodding, J.; Minigo, G.; Pouniotis, D. S.;
Plebanski, M.;Scholzen, A.; McKenzie, I. F. C.; Pietersz, G. A.;
Apostolopoulos, V.Immunology 2007, 120,
325–335.doi:10.1111/j.1365-2567.2006.02506.x
30. Apostolopoulos, V.; McKenzie, I. F.; Pietersz, G. A.Immunol.
Cell Biol. 1996, 74, 457–464. doi:10.1038/icb.1996.76
31. Tzakos, A. G.; Briasoulis, E.; Thalhammer, T.; Jager,
W.;Apostolopoulos, V. J. Drug Delivery 2013, No.
918304.doi:10.1155/2013/918304
32. Kapoor, P.; Singh, H.; Gautam, A.; Chaudhary, K.; Kumar,
R.;Raghava, G. P. S. PLoS One 2012, 7,
e35187.doi:10.1371/journal.pone.0035187
33. Carson-Jurica, M. A.; Schrader, W. T.; O'Malley, B. W.
Endocr. Rev.1990, 11, 201–220. doi:10.1210/edrv-11-2-201
34. Su, H.; Koo, J. M.; Cui, H. J. Controlled Release 2015, 219,
383–395.doi:10.1016/j.jconrel.2015.09.056
35. Apostolopoulos, V.; Deraos, G.; Matsoukas, M.-T.; Day,
S.;Stojanovska, L.; Tselios, T.; Androutsou, M.-E.; Matsoukas,
J.Bioorg. Med. Chem. 2017, 25,
528–538.doi:10.1016/j.bmc.2016.11.029
36. Lourbopoulos, A.; Deraos, G.; Matsoukas, M.-T.; Touloumi,
O.;Giannakopoulou, A.; Kalbacher, H.; Grigoriadis,
N.;Apostolopoulos, V.; Matsoukas, J. Bioorg. Med. Chem. 2017,
25,4163–4174. doi:10.1016/j.bmc.2017.06.005
37. Hou, J.; Diao, Y.; Li, W.; Yang, Z.; Zhang, L.; Chen, Z.;
Wu, Y.Int. J. Pharm. 2016, 505, 329–340.
doi:10.1016/j.ijpharm.2016.04.017
38. Fung, S.; Hruby, V. J. Curr. Opin. Chem. Biol. 2005, 9,
352–358.doi:10.1016/j.cbpa.2005.06.010
39. Johnson, M.; Liu, M.; Struble, E.; Hettiarachchi, K.J.
Pharm. Biomed. Anal. 2015, 109,
112–120.doi:10.1016/j.jpba.2015.01.009
40. Pierschbacher, M. D.; Ruoslahti, E. Nature 1984, 309,
30–33.doi:10.1038/309030a0
41. Kapp, T. G.; Rechenmacher, F.; Neubauer, S.; Maltsev, O.
V.;Cavalcanti-Adam, E. A.; Zarka, R.; Reuning, U.; Notni,
J.;Wester, H.-J.; Mas-Moruno, C.; Spatz, J.; Geiger, B.; Kessler,
H.Sci. Rep. 2017, 7, No. 39805. doi:10.1038/srep39805
42. Schwartz, M. A.; Schaller, M. D.; Ginsberg, M. H.Annu. Rev.
Cell Dev. Biol. 1995, 11,
549–599.doi:10.1146/annurev.cb.11.110195.003001
43. Plow, E. F.; Haas, T. A.; Zhang, L.; Loftus, J.; Smith, J.
W.J. Biol. Chem. 2000, 275, 21785–21788.
doi:10.1074/jbc.R000003200
44. Gilad, Y.; Firer, M.; Gellerman, G. Biomedicines 2016, 4,
No. 11.doi:10.3390/biomedicines4020011
45. Cox, D.; Brennan, M.; Moran, N. Nat. Rev. Drug Discovery
2010, 9,804–820. doi:10.1038/nrd3266
46. Chen, K.; Chen, X. Theranostics 2011, 1,
189–200.doi:10.7150/thno/v01p0189
47. Kumar, C. C. Curr. Drug Targets 2003, 4,
123–131.doi:10.2174/1389450033346830
48. Ganguly, K. K.; Pal, S.; Moulik, S.; Chatterjee, A. Cell
Adhes. Migr.2013, 7, 251–261. doi:10.4161/cam.23840
49. Liu, Z.; Wang, F.; Chen, X. Drug Dev. Res. 2008, 69,
329–339.doi:10.1002/ddr.20265
50. Cai, W.; Chen, X. Anti-Cancer Agents Med. Chem. 2006, 6,
407–428.doi:10.2174/187152006778226530
51. Burkhart, D. J.; Kalet, B. T.; Coleman, M. P.; Post, G. C.;
Koch, T. H.Mol. Cancer Ther. 2004, 3, 1593–1604.
52. Gilad, Y.; Noy, E.; Senderowitz, H.; Albeck, A.; Firer, M.
A.;Gellerman, G. Pept. Sci. 2016, 106, 160–171.
doi:10.1002/bip.22800
53. Baba, Y.; Matsuo, H.; Schally, A. V.Biochem. Biophys. Res.
Commun. 1971, 44, 459–463.doi:10.1016/0006-291X(71)90623-1
54. Marelli, M. M.; Moretti, R. M.; Januszkiewicz-Caulier, J.;
Motta, M.;Limonta, P. Curr. Cancer Drug Targets 2006, 6,
257–269.doi:10.2174/156800906776842966
55. Limonta, P.; Marelli, M. M.; Mai, S.; Motta, M.; Martini,
L.;Moretti, R. M. Endocr. Rev. 2012, 33,
784–811.doi:10.1210/er.2012-1014
https://doi.org/10.1038%2Fnrd1470https://doi.org/10.1038%2Fnrd3504https://doi.org/10.1038%2Fnrd4145https://doi.org/10.1038%2Fnrd4505https://doi.org/10.1038%2Fnrd.2016.256https://doi.org/10.1016%2Fj.jconrel.2010.04.009https://doi.org/10.3402%2Fnano.v3i0.18496https://doi.org/10.1158%2F0008-5472.CAN-15-2138https://doi.org/10.1021%2Fbc500081ghttps://doi.org/10.1039%2FC7OB02045Ghttps://doi.org/10.1126%2Fsciadv.1600200https://doi.org/10.1038%2Fnrd2803https://doi.org/10.1186%2F1475-2867-13-89https://doi.org/10.1016%2Fj.yexmp.2008.12.004https://doi.org/10.2174%2F1385272820666151211192249https://doi.org/10.1016%2Fj.ijpharm.2016.07.008https://doi.org/10.1016%2Fj.ejpb.2013.12.009https://doi.org/10.1186%2Fbcr1505https://doi.org/10.1111%2Fj.1365-2567.2006.02506.xhttps://doi.org/10.1038%2Ficb.1996.76https://doi.org/10.1155%2F2013%2F918304https://doi.org/10.1371%2Fjournal.pone.0035187https://doi.org/10.1210%2Fedrv-11-2-201https://doi.org/10.1016%2Fj.jconrel.2015.09.056https://doi.org/10.1016%2Fj.bmc.2016.11.029https://doi.org/10.1016%2Fj.bmc.2017.06.005https://doi.org/10.1016%2Fj.ijpharm.2016.04.017https://doi.org/10.1016%2Fj.cbpa.2005.06.010https://doi.org/10.1016%2Fj.jpba.2015.01.009https://doi.org/10.1038%2F309030a0https://doi.org/10.1038%2Fsrep39805https://doi.org/10.1146%2Fannurev.cb.11.110195.003001https://doi.org/10.1074%2Fjbc.R000003200https://doi.org/10.3390%2Fbiomedicines4020011https://doi.org/10.1038%2Fnrd3266https://doi.org/10.7150%2Fthno%2Fv01p0189https://doi.org/10.2174%2F1389450033346830https://doi.org/10.4161%2Fcam.23840https://doi.org/10.1002%2Fddr.20265https://doi.org/10.2174%2F187152006778226530https://doi.org/10.1002%2Fbip.22800https://doi.org/10.1016%2F0006-291X%2871%2990623-1https://doi.org/10.2174%2F156800906776842966https://doi.org/10.1210%2Fer.2012-1014
-
Beilstein J. Org. Chem. 2018, 14, 930–954.
952
56. Chen, A.; Kaganovsky, E.; Rahimipour, S.; Ben-Aroya, N.;
Okon, E.;Koch, Y. Cancer Res. 2002, 62, 1036–1044.
57. Cheung, L. W. T.; Yung, S.; Chan, T.-M.; Leung, P. C.
K.;Wong, A. S. T. Mol. Ther. 2013, 21, 78–90.
doi:10.1038/mt.2012.187
58. Padula, A. M. Anim. Reprod. Sci. 2005, 88,
115–126.doi:10.1016/j.anireprosci.2005.05.005
59. Zhu, S.; Wang, Q.; Jiang, J.; Luo, Y.; Sun, Z. Sci. Rep.
2016, 6,No. 33894. doi:10.1038/srep33894
60. Dharap, S. S.; Wang, Y.; Chandna, P.; Khandare, J. J.; Qiu,
B.;Gunaseelan, S.; Sinko, P. J.; Stein, S.; Farmanfarmaian, A.;
Minko, T.Proc. Natl. Acad. Sci. U. S. A. 2005, 102,
12962–12967.doi:10.1073/pnas.0504274102
61. Schally, A. V.; Nagy, A. Eur. J. Endocrinol. 1999, 141,
1–14.doi:10.1530/eje.0.1410001
62. Laimou, D.; Katsila, T.; Matsoukas, J.; Schally, A.;
Gkountelias, K.;Liapakis, G.; Tamvakopoulos, C.; Tselios, T. Eur.
J. Med. Chem.2012, 58, 237–247.
doi:10.1016/j.ejmech.2012.09.043
63. Katsila, T.; Balafas, E.; Liapakis, G.; Limonta, P.;
Marelli, M. M.;Gkountelias, K.; Tselios, T.; Kostomitsopoulos, N.;
Matsoukas, J.;Tamvakopoulos, C. J. Pharmacol. Exp. Ther. 2011, 336,
613–623.doi:10.1124/jpet.110.174375
64. Martinez, V. Chapter 180 - Somatostatin A2. In Handbook
ofBiologically Active Peptides, 2nd ed.; Kastin Abba, J., Ed.;
AcademicPress: Boston, 2013; pp
1320–1329.doi:10.1016/B978-0-12-385095-9.00180-9
65. Keskin, O.; Yalcin, S. OncoTargets Ther. 2013, 6,
471–483.doi:10.2147/OTT.S39987
66. Lahlou, H.; Guillermet, J.; Hortala, M.; Vernejoul, F.;
Pyronnet, S.;Bousquet, C.; Susini, C. Ann. N. Y. Acad. Sci